The MIT Computer Science and Artificial Intelligence Laboratory (CSAIL) program has come up with some amazing advances in robotics recently, from origami robots that transform themselves to artificial intelligence that can sense people through walls. Its newest project allows you to control a robot just by watching it and correct mistakes with a simple hand gesture.
The team demonstrated the results of their research with a short video showing a human supervising a robot drilling holes in a piece of wood. The interface works on people the robot has never encountered before, meaning there’s no training involved.
The brain sensors can quickly detect when a person notices that the robot is about to make a mistake. Using hand movement, the robot can then be instructed the correct action to perform. CSAIL Director Daniela Rus said the two sensors working in tandem enabled an almost instantaneous response.
“This work combining EEG and EMG feedback enables natural human-robot interactions for a broader set of applications than we’ve been able to do before using only EEG feedback,” Rus said. “By including muscle feedback, we can use gestures to command the robot spatially, with much more nuance and specificity.”
Controlling a robot with your brain often requires you to learn how to “think” in a certain way so the sensors can interpret the commands correctly. It’s one thing in a controlled laboratory environment with a trained operator, but you can imagine how it might be difficult on a noisy construction site, for example.
Everyone’s familiar with the “uh-oh” moment you get when you realize something is about to go haywire. For this project, the brain-wave scanners could quickly detect signals known as “error-related potentials” (ErrPs), which occur naturally when people notice mistakes. An ErrP causes the robot to pause, so the human operator can direct the operation correctly, if needed.
“What’s great about this approach is that there’s no need to train users to think in a prescribed way,” Joseph DelPreto, lead author of a paper on the research, said.“The machine adapts to you, and not the other way around.”
In the study, Baxter the robot chose the correct drill spot 70 percent of the time on his own. With human supervision, that number rose to 97 percent.
“By looking at both muscle and brain signals, we can start to pick up on a person’s natural gestures along with their snap decisions about whether something is going wrong,” DelPreto said. “This helps make communicating with a robot more like communicating with another person.”
San Francisco startup Dimensional advertised for people to walk robot dogs through city streets. The New York Times reported that the job could include carrying a roughly 33-pound robot back to the office if it overheated. A lift home is an awkward requirement for a labor-saving machine.
Keeping the hardware running is only one challenge. Anthropic's robotics tests found that language models mostly failed when controlling joints directly, but did better when directing trained controllers. Giving an AI agent a body doesn't teach it to handle ordinary household objects.
Soon, a mattress protector may be more than a bed cover. Australian medtech company Sleeptite and researchers at RMIT University have received a US patent for FlexiSense, a smart bedding technology that tracks body position and posture to help caregivers identify when older residents may be at risk of falling.
The technology is being developed for Sleeptite’s in-bed monitoring system, REMi, and was recently on display at The MedTech Conference in Boston. A peer-reviewed study in Flexible and Printed Electronics says the sensors can also be made at scale without sacrificing performance or durability.
AI companies spend enormous amounts of time convincing us their technology is safe and secure. But what happens when those same companies become targets of increasingly sophisticated AI-powered attacks? Anthropic, the company behind Claude, is now facing that question. According to a Washington Post report, cybersecurity startup Cathedral discovered vulnerabilities that let its researchers access sensitive information in Anthropic's internal systems.
The researchers used AI agents to identify potential entry points in less than 12 hours. Interestingly, they didn't need to crack Claude itself. Instead, they exploited weaknesses in third-party business applications that Anthropic uses for everyday operations. Anthropic's security testing program disclosed the findings, which allows outside researchers to identify and report vulnerabilities.